JP2013529784A - Method for detecting cancer - Google Patents

Abstract

Disclosed herein are methods for determining the absence or presence of one or more cancer types in an animal. In some embodiments, the amount of lipid in a sample (eg, body fluid or processed product thereof) from the animal is used with a predictive model to determine. The amount of lipid can be measured using a mass spectrometry system in some cases. In one embodiment, a method is provided for determining the presence or absence of at least one cancer type in an animal, the method determining the lipid content of the lipid in a lipid set in a sample derived from the animal. Determining the presence or absence of at least one cancer type in the animal with a prediction model, wherein the lipid content of lipids in the lipid set includes the input of the prediction model And the sample contains bodily fluids or processed products thereof.

Description

(Cross-reference to related applications)
This application claims the benefit of US Provisional Application No. 61 / 357,642, filed June 23, 2010, which is hereby incorporated by reference in its entirety.

(Government rights)
This invention was made in part with government support under grant number EPS-0447479 awarded by the National Science Foundation / Office of Experimental Program to Stimulate Competitive Research (EPSCoR). The US government has certain rights in the invention.

(background)
Cancer is a major cause of death, refuge, and a major damage to the medical system in the United States and throughout the world. Early detection is associated with better treatment options and improved results. Thus, early detection of cancer can help minimize both hardship and cost while typically increasing the chance of survival. Accordingly, some embodiments of the invention are methods for determining the presence of cancer in an animal.

(Summary)
Embodiments of the invention include a method for determining the presence or absence of at least one cancer type in an animal, said method comprising a lipid set in a sample from said animal. determining the lipid amount of the lipid in set) and determining the presence or absence of at least one cancer type in the animal with a predictive model. In some of these methods, the lipid content of the lipid in the lipid set includes the input of the prediction model, and the sample includes body fluid or treatment thereof.

  In some embodiments, the bodily fluid comprises plasma vomiting, earwax, gastric juice, breast milk, mucus, saliva, sebum, semen, sweat, tears, vaginal discharge, serum, aqueous humor, vitreous humor, Group consisting of endolymph fluid, perilymph fluid, peritoneal fluid, pleural fluid, cerebrospinal fluid, blood, plasma, nipple aspirate fluid, urine, stool, and bronchoalveolar fluid fluid More selected. In yet another embodiment, the body fluid is blood or plasma. In still other embodiments, the sample comprises a lipid microvesicle fraction.

  In some exemplary embodiments, the lipid set comprises at least 10 lipids, at least 50 lipids, at least 100 lipids, at least 200 lipids, or no more than 100,000 lipids.

  In some cases, the lipid set is BMP, CE, Cer, DAG, DH-LTB4, FA, GA2, GM3, HexCer, HexDHCer, LacCer, LysoPA, LysoPC, LysoPC-pgg, LysoPE, LysoPE-pmg, LysoPS. Selected from one or more classes of lipids selected from the group consisting of: MAG, PC, PC-pmg, PE, PE-pmg, PGA1, PGB1, SM, sphingosine, TAG, and TH-12-keto-LTB4 One or more lipids. In other cases, the lipid set is

One or more lipids selected from one or more classes of lipids selected from the group consisting of: In yet other cases, the one or more lipids in the lipid set are:

Selected from the group consisting of

  In some embodiments, the at least one cancer type comprises lung cancer, and the one or more lipids in the lipid set are LysoPA (22: 0), PE-pmg (42: 9), FA (16 : 3), FA (19: 1), CE (18: 2), Cer (36: 1), Cer (38: 4), PC (38: 5), Cer (38: 1), and TAG (44 : Selected from the group consisting of 3). In still other embodiments, the at least one cancer type comprises lung cancer, and the one or more lipids in the lipid set are:

Selected from the group consisting of In a further embodiment, the at least one cancer type comprises lung cancer and the one or more lipids in the lipid set are TAG (44: 3), PC (36: 5), PC (38: 5), Cer (38: 4), PE-pmg (42: 9), PC (38: 7), LysoPA (22: 0), Cer (38: 1), Cer (34: 1), and Cer (36: 1) ).

  In some embodiments of the invention, the at least one cancer type comprises breast cancer, and the one or more lipids in the lipid set are LysoPA (22: 1), PE-pmg (42: 9), CE (20: 5), TAG (52: 3), LysoPA (22: 0), PC (36: 3), PC (36: 4), PC (36: 2), PC (34: 2), and Selected from the group consisting of PC (34: 1). In another embodiment, the at least one cancer type comprises breast cancer, and the one or more lipids in the lipid set are:

Selected from the group consisting of In still other embodiments, the at least one cancer type comprises breast cancer, and the one or more lipids in the lipid set are PC (34: 2), PC (34: 1), PC (36: 2 ), PC (36: 4), PC (36: 3), PC (38: 4), LysoPA (22: 1), PE-pmg (42: 9), LysoPA (22: 0), and CE (20 : Selected from the group consisting of 5).

  In some embodiments of the invention, the at least one cancer type comprises lung cancer and breast cancer, and the one or more lipids in the lipid set are LysoPA (22: 1), PC (36: 5), TAG (52: 3), PC (38: 5), CE (20: 5), TAG (50: 2), BMP (39: 1), PC (34: 2), CE (18: 2), and Selected from the group consisting of PC (34: 1). In other embodiments, the at least one cancer type comprises lung cancer and breast cancer, and the one or more lipids in the lipid set are:

Selected from the group consisting of In still other embodiments, the at least one cancer type comprises lung cancer and breast cancer, and the one or more lipids in the lipid set is PC (34: 2), PC (36: 2), TAG (44 : 3), CE (18: 2), PC (34: 1), LysoPA (22: 1), PC (36: 5), Cer (36: 1), CE (20: 5), and PC (36 : Selected from the group consisting of 3).

  In some embodiments of the invention, the amount of lipid is determined using mass spectrometry (eg, with a Fourier transform ion cyclotron resonance mass spectrometer).

  In another embodiment, the sample is a processed body fluid. For example, the sample is a processed body fluid containing one or more extracts using one or more solutions containing acetonitrile, water, chloroform, methanol, butylhydroxytoluene, trichloroacetic acid, or combinations thereof. possible.

  In some embodiments, the predictive model may include one or more dimension reduction methods (eg, principal component analysis (PCA), soft independence modeling of class). SIMS), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OP) consisting of an OP consisting of an orthogonal partial least squares discriminant analysisDA One or more methods).

  In some embodiments of the invention, the animal is selected from the group consisting of humans, dogs, cats, horses, cows, pigs, sheep, chickens, turkeys, mice, and rats.

  In another embodiment of the present invention, the at least one cancer type is carcinoma, sarcoma, blood cancer, neurological malignancy, basal cell cancer, thyroid cancer, neuroblastoma, ovarian cancer, melanoma, kidney Cell carcinoma, hepatocellular carcinoma, breast cancer, colon cancer, lung cancer, pancreatic cancer, brain cancer, prostate cancer, chronic lymphocytic leukemia, acute lymphoblastic leukemia, rhabdomyosarcoma, glioblastoma multiforme, meninges Tumor, bladder cancer, gastric cancer, glioma, oral cancer, nasopharyngeal cancer, kidney cancer, rectal cancer, lymph node cancer, bone marrow cancer, stomach cancer, uterine cancer, Selected from the group consisting of leukemia, basal cell carcinoma, cancers related to epithelial cells, cancers capable of altering the regulation or activity of pyruvate carboxylase, and tumors associated with any of the aforementioned cancer types.

  In a further embodiment, the method includes determining the presence or absence of more than one cancer type.

  Embodiments of the invention include a method for determining the presence or absence of at least one cancer type in an animal, the method comprising the presence or absence of at least one cancer type in the animal, It includes the step of determining with a predictive model by analyzing the lipid content of lipids in a lipid set in a sample derived from the animal. In some cases, the lipid content of lipids in the lipid set includes the input of the prediction model, and the sample includes body fluids or processed products thereof.

  The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The invention may be better understood by reference to one or more of these drawings in combination with the description of specific embodiments presented herein.

FIG. 1 is diacylglycerol in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 2 is phosphatidylcholine in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 3 is phosphatidylcholine in samples from humans with lung cancer, breast cancer, and no cancer (control). FIG. 4 is phosphatidylcholine in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 5 is phosphatidylcholine in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 6 is phosphatidylethanolamine in samples from humans with lung cancer, breast cancer, and no cancer (control). FIG. 7 is phosphatidylethanolamine in samples from humans with lung cancer, breast cancer, and no cancer (control). FIG. 8 is phosphatidylethanolamine in samples from humans with lung cancer, breast cancer, and no cancer (control). FIG. 9 is phosphatidylethanolamine-pmg in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 10 is phosphatidylethanolamine-pmg in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 11 is triacylglycerol in samples from lung cancer, humans with breast cancer, and humans without cancer (control). FIG. 12 is monoacylglycerol in samples from humans with lung cancer, breast cancer and no cancer (control). FIG. 13 is a lipid class in a sample derived from a human comparing cancer types. FIG. 14 shows PC1-PC3 PCA score plot-4 components, R ² X = 0.475, Q ² = 0.296 for breast cancer, control, and lung cancer samples. FIG. 15 shows PC1-PC3 PCA loading plot-4 components of breast cancer, control, and lung cancer samples, R ² X = 0.475, Q ² = 0.296. FIG. 16 is an OPLS-DA score plot that separates lung cancer and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.253, R} 2 Y = 0.619, Q 2 = 0.345. FIG. 17 is an OPLS-DA coefficient plot that separates lung cancer and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.253, R} 2 Y = 0.619, Q 2 = 0.345. The coefficient plot shows elevated lipids in the control sample. FIG. 18 is an OPLS-DA coefficient plot that separates lung cancer and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.253, R} 2 Y = 0.619, Q 2 = 0.345. The coefficient plot shows elevated lipids in lung cancer samples. FIG. 19 is an OPLS-DA score plot separating breast and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.281, R} 2 Y = 0.762, Q 2 = 0.625. FIG. 20 is an OPLS-DA coefficient plot that separates breast cancer and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.281, R} 2 Y = 0.762, Q 2 = 0.625. The coefficient plot shows elevated lipids in the control sample. FIG. 21 is an OPLS-DA coefficient plot that separates breast cancer and control samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.281, R} 2 Y = 0.762, Q 2 = 0.625. The coefficient plot shows elevated lipids in breast cancer samples. FIG. 22 is an OPLS-DA score plot that separates lung and breast cancer samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.309, R} 2 Y = 0.816, Q 2 = 0.725. FIG. 23 is an OPLS-DA coefficient plot that separates lung and breast cancer samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.309, R} 2 Y = 0.816, Q 2 = 0.725. The coefficient plot shows elevated lipids in lung cancer samples. FIG. 24 is an OPLS-DA coefficient plot that separates lung and breast cancer samples. - 1 + 1 orthogonal ^{component, R 2 X = 0.309, R} 2 Y = 0.816, Q 2 = 0.725. The coefficient plot shows elevated lipids in breast cancer samples. FIG. 25 is a flow chart representative of one embodiment of the methods disclosed herein. A cancer condition can include the presence or absence of one or more cancer types.

(Detailed explanation)
Some embodiments of the invention include a method for detecting the presence or absence of one or more cancer types by determining the amount of lipid in a set of lipids in a sample. The sample may be a body fluid derived from an animal (or a processed product thereof). In some cases, the sample (eg, bodily fluid extract) contains a higher concentration of lipid microvesicles than is normally found in bodily fluids. The amount of lipid in the lipid set is analyzed using a predictive model to determine the presence or absence of one or more cancer types.

  Lipids are indicated according to the following notation XXX (YY: ZZ). XXX is an abbreviation for the lipid group (in many cases indicating a lipid head group), for example, as provided in Table 1. YY is the number of carbon atoms in the acyl chain. ZZ is the number of double bonds in the acyl chain.

  The term lipid, as used herein, is defined as a collection of one or more isomers. For example, PC (36: 1) is a lipid and one of the phosphatidylcholine isomers that has 36 carbons in the acyl chain and one double bond in either of the two acyl chains. More than one species; these isomers have the same molecular weight. The term lipid can encompass an entire collection of isomers, but the sample is actually only one isomer, several isomers, or any number less than the total number of all possible isomers in the collection. Of isomers. Thus, a lipid may refer to one or more of the isomers that make up the entire collection of possible isomers.

  The term lipid set is defined to include one or more lipids.

  The term lipid amount (and similar phrases (eg, amounts of lipids or amount of a lipid)) is the absolute amount of lipid (eg, in mmole) or the relative amount of lipid (eg,% relative In units of intensity).

  Table 1 provides the lipid name abbreviations used in the data; other abbreviations are provided in the text if required.

Body fluid is any suitable body fluid for cancer determination, including vomiting, earwax (ear ears), gastric fluid, breast milk, mucus (eg, runny nose and sputum), saliva, sebum (skin oil), semen (prostate gland) Fluid), sweat, tears, vaginal discharge, serum, aqueous humor, vitreous humor, endolymph fluid, perilymph fluid, peritoneal fluid, pleural fluid, cerebrospinal fluid, blood, plasma, nipple aspirate fluid, urine, feces , Bronchoalveolar lavage fluid, peripheral blood, serum, plasma, ascites, cerebrospinal fluid (CSF), sputum, bone marrow, synovial fluid, aqueous humor, amniotic fluid, Cowper's gland fluid or urethral gland fluid, female ejection fluid, feces (Fetal matter), hair, cyst fluid, pleural fluid and peritoneal fluid, pericardial fluid, lymph, sputum, milk fistula, bile, interstitial fluid, menstrual discharge, pus, vaginal discharge, mucosal discharge, watery stool (Tool water), pancreatic juice, lavage fluid, bronchopulmonary aspirate, or others Cleaning liquid including but not limited to. Body fluids can also include maternal circulation, which can be blastocyst cavity, umbilical cord blood, or fetal or maternal origin.

  The bodily fluid can be obtained from any animal tissue (eg, mammalian tissue) and includes, but is not limited to, connective tissue, muscle tissue, nerve tissue, adipose tissue, endothelial tissue, or epithelial tissue. The tissue can be at least part of an organ or part of an organ system. Organs include heart, blood, blood vessels, salivary glands, esophagus, stomach, liver, gallbladder, pancreas, large intestine, small intestine, rectum, anus, colon, endocrine glands (eg, hypothalamus, pituitary gland, pineal gland, thyroid gland, accessory gland) Thyroid and adrenal gland), kidney, ureter, bladder, urethrakin, hair, nails, lymph, lymph nodes, lymphatic vessels, leukocytes, tonsils, pharyngeal tonsils, thymus, spleen, muscles, brain, spinal cord, peripheral nerves, Nerves, genitals (eg, ovary, fallopian tube, uterus, vagina, mammary gland (eg, breast), testis, vagina, seminal vesicle, prostate and penis), pharynx, larynx, trachea, bronchi, lung, diaphragm, bone, cartilage , Ligaments, or tendons, but are not limited to these. The organ system may include, but is not limited to, the circulatory system, digestive system, endocrine system, excretory system, integumental system, lymphatic system, muscular system, nervous system, genital system, respiratory system, or skeletal system. Not.

  Body fluids can be collected by any suitable method (blood collection, natural drainage manipulation (eg, by suction or by manual manipulation (eg, nipple to obtain nipple aspirate)), surgical method (eg, excision) , Can be removed from the animal by biopsy methods (eg, but not limited to, puncture aspiration or needle biopsy), or organ removal and dissection after animal sacrifice. The removed body fluid can be frozen in liquid nitrogen. Preparation of the removed body fluid can be done in any suitable manner.

  The body fluid may be obtained through a third party (eg, a person who does not perform analysis). For example, the bodily fluid can be obtained through a clinician, doctor, or other healthcare manager of the subject from which the sample is derived. In some embodiments, the body fluid may be obtained by the same party performing the analysis.

  In some embodiments, the bodily fluid is the sample. In other embodiments, the bodily fluid is processed to provide the sample. Processing includes any suitable method (extraction, centrifugation (eg, ultracentrifugation), lyophilization, fractionation, separation (eg, using column or gel chromatography), or evaporation. Without limitation). In some cases, the treatment includes any suitable solvent or combination of solvents (eg, acetonitrile, water, chloroform, methanol, butylhydroxytoluene, trichloroacetic acid, toluene, hexane, benzene, or combinations thereof. May include one or more extractions with a solution including, but not limited to. For example, in some embodiments, the blood-derived fraction is extracted with a mixture comprising methanol and butylhydroxytoluene. In some cases, the sample (eg, a bodily fluid extract or plasma lipid microvesicle fraction) contains a higher concentration of lipid microvesicles than is normally found in bodily fluids.

  The volume of the sample used for analysis (eg, the body fluid or processed product thereof) ranges from about 0.1 mL to about 20 mL (eg, about 20 mL or less, about 15 mL, about 10 mL, about 9 mL, about 8 mL). About 7 mL, about 6 mL, about 5 mL, about 4 mL, about 3 mL, about 2 mL, about 1 mL, or about 0.1 mL).

  A broad class of lipids that can be part of a lipid set includes, but is not limited to: fatty acids (characterized by carboxyl groups and acyl chains), glycerolipids (glycerol backbone and one ( Monoacylglycerol (MAG)), two (diacylglycerol (DAG)) or three (triacylglycerol (TAG)) ester-linked fatty acyl chains), glycerophospholipid (GPL) ) (Characterized by a glyceryl skeleton with two ester-linked acyl chains and a phosphate-linked polar head group, GPLs include phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylserine (PS) Phosphatidylglycero (PG), phosphatidylinositol (PI), inositol, and phosphatidic acid), lysoglycerophospholipid (LGPL) (LGPL lacks one of the acyl chains in the glycerol backbone (eg, at the C2 position). LGPL includes lysophosphatidylcholine (LysoPC), lysophosphatidylethanolamine (LysoPE), lysophosphatidylserine (LysoPS), lysophosphatidylglycerol (LysoPG), lysophosphatidylinositol (LysoPI), lysophosphatidic acid and lysophosphatidic acid. Sphingolipid (SPL) (trans between sphingosine backbone and C4 and C5 of acyl groups linked to amino groups via amide bonds) Characterized by a double bond), bis (monoacylglycero) phosphate (BMP), ceramides, gangliosides, sterols, prenol, Saccharomyces lipids (Saccharolipid), and polyketides. In some cases, the lipid group is based on the composition of the acyl chain, which can vary in any number of ways, including the number of carbons in the acyl chain and the number of double bonds in the acyl chain. .

  In other embodiments, the lipids in the lipid set can be derived from one or more classes of lipids (eg, BMP, CE, Cer, DAG, DH-LTB4, FA, GA2, GM3, HexCer, HexDHCer). , LacCer, LysoPA, LysoPC, LysoPC-pmg, LysoPE, LysoPE-pmg, LysoPS, MAG, PC, PC-pmg, PE, PE-pmg, PGA1, PGB1, SM, sphingosine, TAG, or TH-12-keto LTB4). In still other embodiments, the lipids in the lipid set can be derived from one or more classes of lipids (eg, FA, MAG, DAG, TAG, PC, PE, PS, PI, PG, PA, LysoPC, LysoPE, LysoPS, LysoPI, LysoPG, LysoPA, LysoPC, LysoPE, BMP, SM, Cer, Cer-P, HexCer, GA1, GA2, GD1, GD2, GM1, GM2, GM3, GT, or GM3, GT).

  In some embodiments, the lipid set used in the prediction model is limited to those more likely to be found in a human lipid pool. In other embodiments, the lipid set excludes very short and very long acyl chains (eg, less than 10 carbons and more than 26 carbons in the chain). In yet other embodiments, the lipids of the lipid set (eg, GPL or LGPL) are limited to those containing an even number of carbons in the acyl chain. In other embodiments, the lipid set included one or more lipids listed in Table 2.

In some embodiments, the lipids in the lipid set can include one or more of the following:

.

  In some embodiments (eg, to determine lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine lung cancer), the lipids in the lipid set include, but are not limited to: LysoPA (22: 0), PE-pmg (42: 9), FA (16: 3), FA (19: 1), CE (18: 2), Cer (36: 1), Cer (38: 4), PC (38: 5), Cer (38: 1), or TAG (44: 3).

  In some embodiments (eg, to determine lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine lung cancer), the lipids in the lipid set include, but are not limited to: TAG (44: 3), PC (36: 5), PC ( 38: 5), Cer (38: 4), PE-pmg (42: 9), PC (38: 7), LysoPA (22: 0), Cer (38: 1), Cer (34: 1), or Cer (36: 1).

  In some embodiments (eg, to determine breast cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast cancer), the lipids in the lipid set include, but are not limited to: LysoPA (22: 1), PE-pmg (42: 9), CE (20: 5), TAG (52: 3), LysoPA (22: 0), PC (36: 3), PC (36: 4), PC (36: 2), PC (34: 2), or PC (34: 1).

  In some embodiments (eg, to determine breast cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast cancer), the lipids in the lipid set include, but are not limited to: PC (34: 2), PC (34: 1), PC ( 36: 2), PC (36: 4), PC (36: 3), PC (38: 4), LysoPA (22: 1), PE-pmg (42: 9), LysoPA (22: 0), or CE (20: 5).

  In some embodiments (eg, to determine breast and lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast and lung cancer), the lipids in the lipid set include, but are not limited to: LysoPA (22: 1), PC (36: 5), TAG (52: 3), PC (38: 5), CE (20: 5), TAG (50: 2), BMP (39: 1), PC (34: 2), CE (18: 2), or PC (34: 1).

  In some embodiments (eg, to determine breast and lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast and lung cancer), the lipids in the lipid set include, but are not limited to:

.

  In some embodiments (eg, to determine breast and lung cancer), the lipids in the lipid set include, but are not limited to: PC (34: 2), PC (36: 2), TAG (44: 3), CE (18: 2), PC (34: 1), LysoPA (22: 1), PC (36: 5), Cer (36: 1), CE (20: 5), or PC (36: 3).

  In some embodiments, the number of lipids in the lipid set may include at least 10, at least 50, at least 100, at least 150, at least 200, at least 500, or at least 1000 lipids. In some embodiments, the number of lipids in the lipid set is 200 or less, 500 or less, 1,000 or less, 5,000 or less, 10,000 or less, or 100,000 or less lipids. Can be included.

  Animals include, but are not limited to, primates (eg, humans), dogs, horses, cows, pigs, sheep, birds, or mammals. Animals include animals as pets or animals in a zoo, including domesticated pigs and horses (including racehorses). In addition, any animal associated with economic activity includes, for example, animals associated with agriculture and aquaculture, and disease monitoring, diagnosis, and treatment options are commonly used in managing economic productivity and / or food chain safety. Animals related to other activities that are common practices. In some embodiments, the animal is a human, dog, cat, horse, cow, pig, sheep, chicken, turkey, mouse, or rat.

  The above cancer types (including cancerous tumors) include carcinoma, sarcoma, blood cancer, neurological malignancy, basal cell carcinoma, thyroid cancer, neuroblastoma, ovarian cancer, melanoma, renal cell carcinoma, liver Cell cancer, breast cancer, colon cancer, lung cancer, pancreatic cancer, brain cancer, prostate cancer, chronic lymphocytic leukemia, acute lymphoblastic leukemia, rhabdomyosarcoma, glioblastoma multiforme, meningioma, bladder cancer , Stomach cancer, glioma, oral cancer, nasopharyngeal cancer, kidney cancer, rectal cancer, lymph node cancer, bone marrow cancer, stomach cancer, uterine cancer, leukemia, basal cell cancer, cancer of epithelial cells, or pyruvate carboxylase Cancers that can alter regulation or activity can be mentioned, but are not limited to these. Examples of cancerous tumors include tumors associated with any of the above cancers.

  In some embodiments, determining the presence or absence of one or more cancer types includes determining the presence or absence of each cancer type.

  The amount of lipid can be determined using any suitable technique, including, for example, any of the mass spectrometry methods described herein.

  A mass spectrometry spectrum of the sample is obtained using a mass spectrometry system. The mass spectrometry system includes the usual components of a mass spectrometer (eg, ionization source, ion detector, mass analyzer, vacuum chamber, and pumping system) and other components (separation systems (eg, interfaced) Chromatographic system), but not limited thereto. The mass spectrometer can be any mass spectrometer suitable for determining lipid content. The mass analyzer system may include any suitable system, including a time-of-flight analyzer, a quadrupole analyzer, a magnetic sector, an Orbitrap, a linear ion trap, or a Fourier transform ion cyclotron resonance (FTICR). , But not limited to). In some cases, the mass analyzer system (eg, FTICR) has a sufficient solution to determine lipid density without further experimental means. The ion sources include electron impact ionization, electrospray, chemical ionization, photoionization, atmospheric pressure chemical ionization, impact ionization, natural ionization, thermal ionization, infarct atom collision ionization, particle beam ionization, or matrix-assisted laser desorption ionization. (MALDI) may be mentioned, but is not limited thereto. In some cases, the electrospray can be a non-ionized ion source. The interfaced chromatographic system may include any suitable chromatographic system (gas chromatography (GC), liquid chromatography (LC), or ion mobility (this may be LC and GC methods). Can be combined), but not limited to. In some cases, direct introduction can be used. In some cases, the mass spectrometry system is GC / MS or LC / MS.

  In some cases, once the MS spectrum is obtained, the spectrum can be analyzed to determine the identity and amount (eg, presence) of lipids in the sample.

  With respect to the MS spectrum, the analysis can include any analysis suitable for determining the identity and / or amount of lipid, including analysis of one or more features, including comparing the mass to a known mass. , Chromatographic retention time (eg, for GC / MS or LC / MS), and mass fragmentation pattern. In some cases, the analysis may include a comparison of features with features of a database (eg, a standard database). In some cases, PREMISE (Lane, A.N., TW-M. Fan, X. Xie, H.N. Mosley, and RM Higashi, Stable isotope analysis of lipid biosynthesis. mass spectroscopy and NMR Anal. Chim. Acta, 2009. 651: p. 201-208.) can be used to identify lipids. The amount of lipid (eg, relative amount) can be determined by any suitable method, including the response function of the ion detector by reference to a spiked standard or by isotope dilution. Not limited to).

  In some embodiments, the data may be pre-processed before the data is input to the prediction model. The pre-processing method may include one or more of any suitable methods, such as a normalization method or a scaling method. In some embodiments, scaling methods may include centering, unit variance, unit variance without centering, Pareto, and Pareto without centering. It is not limited.

  The predictive model can include any model suitable for determining the presence or absence of one or more cancer types. For example, the prediction model may include one or more dimension reduction methods. In some embodiments, the predictive model may be a clustering method (eg, K-means clustering and K-nearest neighbor clustering), a machine learning method (eg, artificial neural network (ANN)). ) And support vector machines (SVM)), principal component analysis (PCA), shimka method (SIMCA), partial least squares (PLS) regression, orthogonal least squares (OPLS) regression, partial least squares discriminant analysis (PLS) -DA) or one or more of orthogonal partial least squares discriminant analysis (OPLS-DA). If desired, these techniques can include one or more suitable methods (mean centering, median centering, Pareto scaling, unit variance scaling, orthogonal signal correction, integration methods, Differentiation methods, cross-validations, and receiver operating characteristic (ROC) curves). In some embodiments, the prediction model may be developed using a set of training data. Here, the training data is designed to generate a set of applicable parameters and coefficients. In some embodiments, the set of parameters and coefficients can be used to determine whether an animal has cancer and, in some cases, the cancer type. The training data can include negative control data (eg, when no cancer is present in the animal) and positive control data (when one or more cancer types are present in the animal). The training data set can be used to establish a predictive model that can determine two or more cancer types.

  FIG. 25 illustrates several embodiments of the above method. In some cases, once a predictive model is established using a training data set, it is shown that the determination of the presence of one or more cancer types can be made from the predictive model without any training. .

  The methods of the present invention can further include determining protein expression, gene expression, or both proteins or their genes. Any suitable protein (or gene thereof) expression can be determined (pyruvate carboxylase, succinyl CoA synthase, phosphoenolpyruvate carboxylase, transketolase, transaldolase, pyruvate dehydrogenase, dehydrogenase, glutaminase, isocitrate dehydrogenase, α Ketoglutarate dehydrogenase, mitochondrial malate dehydrogenase, succinate dehydrogenase, fumarate hydratase, hexokinase II, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate kinase 1, lactate dehydrogenase 5, phosphofructokinase 1 and 2, glutathione peroxidase Or glutathione-S-transferase, or Proteins associated with the pathway (eg, Krebs cycle (also known as citrate cycle), glycolysis, pentose phosphate pathway (oxidative and non-oxidative), gluconeogenesis, lipid biosynthesis, amino acid synthesis (eg, non-essential amino acids Synthesis), catabolic pathways, urea cycle, coli cycle or glutamic acid / glutamine cycle, but not limited thereto). Protein expression can be determined by gel electrophoresis techniques (eg, Western blotting), chromatography techniques, antibody-based techniques, centrifugation techniques, or techniques including, but not limited to, combinations thereof. Gene expression can include any suitable technique (PCR-based techniques (eg, real-time PCR), gel electrophoresis techniques, chromatography techniques, antibody-based techniques, centrifugation techniques, or combinations thereof) Non-limiting). A method for measuring gene expression can include measuring the amount of cDNA made from tissue isolated RNA.

Blood preparation: Blood was collected from humans diagnosed with breast cancer, lung cancer, humans diagnosed with non-small cell lung cancer (NSCLC), and healthy (cancer free) humans (also referred to as controls or normal). Blood was collected in K-EDTA containing vacutainer tubes (purple lid) for anticoagulation, immediately placed on ice and centrifuged at 3500 × g for 15 minutes at 4 ° C. The supernatant (plasma) was divided into aliquots, then snap frozen in liquid N ₂ and maintained at −80 ° C. until fractions were prepared.

  Fraction preparation: Plasma was thawed and 0.8 ml was transferred to a 1 ml polyallomer ultracentrifuge tube and adjusted to the correct volume using cold PBS (the entire volume of the ultracentrifuge tube is collapsed during centrifugation) To be less than 5 mm from the top). The rotor SWTi55 with a bucket was cooled in advance to 4 ° C. The ultracentrifuge tube and lid in the bucket were weighed and adjusted using PBS to keep all sample mass within a 10 mg variation. The sample was then centrifuged at 70,000 × g (27,172 rpm) for 1 hour at 4 ° C. in a SWTi55 rotor. The supernatant was again centrifuged in SWTi55 at 100,000 × g (32,477 rpm) for 1 hour at 4 ° C. to pellet the lipid microvesicle fraction. The pellet from the first centrifugation (microvesicle fraction) and the lipid microvesicle fraction pellet are washed in cold PBS by resuspension and are in SWTi55 at 100,000 × g (32,477 rpm). ) At 4 ° C. for 1 hour. The supernatant was removed and both tubes were inverted on a paper towel to drain excess PBS. The pellet was resuspended in 2 × 100 μl 18 MOhm water for transfer to a 2 ml microcentrifuge tube and lyophilized overnight. The lyophilized pellet was maintained at −80 ° C. until lipid extraction.

  Lipid extraction: The lipid microvesicle fraction pellet is 0.5 ml methanol (mass spectrometry grade) +1 mM by homogenization with 3 × 3 mm glass beads at 30 Hz for 1 minute in a mixer mill (eg MM200, Retsch). Extracted in butylhydroxytoluene. The homogenate was then shaken for 30 minutes in a shaker and then centrifuged at 14,000 rpm for 30 minutes at 4 ° C. in a microcentrifuge. The supernatant was transferred to a 1.5 ml screw cap glass vial with a Teflon (registered trademark) surface treated silicone septum. Record the extract weight. The lipid extract was stored at −80 ° C. until FT-ICR-MS analysis.

  FT-ICR-MS analysis: Samples were diluted 1: 5 in methanol + 1 mM BHT + 1 ng / nl reserpine and then as previously described (Lane, AN, TW-M. Fan, X. Xie, H. N. Mosley, and R. M. Higashi, Stable isotope analysis of lipid biosynthesis by high resolution mass spectrometry and NMR. Trap-FT-ICR Mass Spectrometer (ThermoFisher LTQ FT, Thermo Electron, Bremen, German) Analysis was performed in y). The FT-ICR-MS was equipped with a TriVersa NanoMate ion source (Advion BioSciences, Ithaca, NY) and an “A” electrospray tip (nozzle inner diameter 5.5 μm). The TriVersa NanoMate was operated by applying 2.0 kV at 0.1 psi head pressure in positive ion mode and 1.5 kV at 0.5 psi in negative ion mode. MS runs were recorded over a mass range of 150-1600 Da. Initially, low resolution MS scans were acquired over 1 minute to ensure ionization stability, after which high mass accuracy data was collected using the FT-ICR analyzer. Here, MS scans were acquired over 8.5 minutes and at a target mass resolution of 200,000 at 400 m / z. AGC (automatic gain control) maximum ion time is set to 500 ms (but typically used at <10 ms) and five “μscans” are acquired for each stored spectrum. Thus, the cycle time for each converted and stored spectrum was about 10 seconds. The LTQ-FT was calibrated and calibrated according to the manufacturer's default standard recommendations, which achieved a mass accuracy better than 1 ppm. The FT-ICR mass spectrum was exported to a spreadsheet file using QualBrowser 2.0 (Thermo Electron) as an accurate mass list (typically exporting all of the observed peaks). Lipid species are first applied with a small (typically <0.0005) linear calibration based on the observed mass of the internal standard (reserpine), and then the in-house software tool PREMISE (PRecalculated Exact Mass Isotopologic Search). Engine (Lane, A.N., T.W.-M. Fan, X. Xie, H.N. Mosley, and R.M.Higashi, Stable isotope analysis of lipid biosynthesis. Chim. Acta, 2009. 651: p. 201-208) (this was verified manually). By using, assigned based on their exact mass. It is customary for PREMISE to be subjected to a selectable window that was less than or equal to 0.0014 Da to match the measured value to the theoretical m / z value. For lipids, the exact masses of a large number (> 3500) of possible GPLs and their ionic forms (exclusively H + and Na + -positive and -H + -negative modes) were precalculated into a spreadsheet lookup table. . The overall method was sufficient to assign GPL to a specific total acyl chain length, degree of saturation, and identity of the head group.

  Chemometric analysis: The normalized FT-ICR-MS data was imported into SimcaP (version 11.5, Umetrics, Umea, Sweden). The above data is averaged and Pareto variance (

). A principal component analysis (PCA) was then performed on the obtained data. We did this and found outlier samples or variables. It was determined that there was no outlier sample and it was a lier when one variable met. The variable sphingosine (18: 1) was excluded from further analysis. The PCA score and loading plot are shown in FIGS.

  An Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA) model was created from three data subsets: normal (ie, no cancer) and lung cancer, normal (ie, no cancer) and breast cancer, and lung cancer and breast cancer. Multivariate from this analysis that maximizes the grouping (eg, the control group and the lung cancer group) while removing one dimension that is orthogonal to the mentioned dimension with the maximum sample variance not related to class separation The dimension in space has been determined. From this analysis, it was determined which of the many variables were most different between classes of data.

  OPLS-DA showed good separation of the total lipid profile from the three classes of plasma. Some of the lipids were essentially the same in each source of lipid microvesicle fraction (general component). The main class that produced the distinction was derived from the loading plot and coefficient plot. These are visualized in the 3D plots shown in FIGS. The OPLS-DA score and coefficient plots are provided in FIGS.

  Lipid seed intensity was normalized to the total lipid response to generate a “mol fraction”. Different lipid classes varied in their amounts. Within some classes, the acyl chain length and the number of double bonds also vary substantially. This was part of the taxonomy. Most of the variance came from objective variability rather than analytical variables. The difference in the amount between classes for distinction was> 4 times and the coefficient of variation within the class was up to 50%. For example, the PC (36: 3) shows the mean and standard deviation of 1.38 ± 0.34 (BrCA) vs. 0.23 ± 0.1 (health) vs. 0.19 ± 0.28 (NSCLC) It was. This example provided statistical separation with p values of <0.0001 (BrCA vs. health), <0.0001 (BrCa vs. NSCLC). NSCLC vs. health did not reach statistical significance. However, other lipids gave a high statistical separation between NSCLC and health. Therefore, several classes were used together to distinguish between healthy individuals and individuals with cancer. The optimal separation is the set of lipids in which at least two of the subject classes differ with a p-value better than 0.01 and at least 10 such lipid classes were used for reliable differentiation. Achieved using.

  Terms such as “preferably”, “generally”, and “typically” may be used to limit the scope of the present invention, whether certain features are critical to the structure or function of the present invention, It also indicates that it is not utilized herein to imply that it is essential or even important. Rather, these terms are merely included to highlight alternative or additional features that may or may not be utilized in a particular embodiment of the present invention.

  A detailed description of one or more embodiments is provided herein. However, it should be understood that the present invention may be embodied in various forms. Accordingly, the specific details disclosed herein (even if indicated as being preferred or advantageous) should not be construed as limiting, but rather as a basis for the claims. And should be used as a representative basis for teaching those skilled in the art to use the invention in any suitable manner.

Claims (31)

A method for determining the presence or absence of at least one cancer type in an animal comprising:
-Determining the lipid content of lipids in a lipid set in a sample derived from said animal; and-determining the presence or absence of at least one cancer type in said animal with a predictive model;
Wherein the lipid content of the lipid in the lipid set comprises the input of the prediction model, and the sample comprises a body fluid or processed product thereof,
Method. The bodily fluids are plasma vomit, earwax, gastric juice, breast milk, mucus, saliva, sebum, semen, sweat, tears, vaginal discharge, serum, aqueous humor, vitreous humor, endolymph, perilymph, peritoneal fluid, pleural fluid The method of claim 1, selected from the group consisting of: cerebrospinal fluid, blood, plasma, nipple aspirate, urine, stool, and bronchoalveolar lavage fluid. The method according to claim 1, wherein the body fluid is blood or plasma. The method of claim 1, wherein the sample comprises a lipid microvesicle fraction. The method of claim 1, wherein the lipid set comprises at least 10 lipids. The method of claim 1, wherein the lipid set comprises at least 50 lipids. The method of claim 1, wherein the lipid set comprises at least 100 lipids. The method of claim 1, wherein the lipid set comprises at least 200 lipids. The method of claim 1, wherein the lipid set comprises 100,000 or fewer lipids. The lipid set is BMP, CE, Cer, DAG, DH-LTB4, FA, GA2, GM3, HexCer, HexDHCer, LacCer, LysoPA, LysoPC, LysoPC-pmg, LysoPE, LysoPE-pmg, LysoPS, MAG, PC, -One or more lipids selected from one or more classes selected from the group consisting of pmg, PE, PE-pmg, PGA1, PGB1, SM, sphingosine, TAG, and TH-12-keto-LTB4 The method of claim 1 comprising. The lipid set is FA, MAG, DAG, TAG, PI, PE, PS, PI, PG, PA, LysoPC, LysoPE, LysoPS, LysoPI, LysoPG, LysoPA, LysoPC, LysoPE, BMP, SM, Cer, Cer-P. 2. One or more lipids selected from one or more classes of lipids selected from the group consisting of: HexCer, GA1, GA2, GD1, GD2, GM1, GM2, GM3, GT1, and CE. The method described in 1. The one or more lipids in the lipid set are:
The method of claim 1, wherein the method is selected from the group consisting of: The at least one cancer type includes lung cancer, and the one or more lipids in the lipid set are LysoPA (22: 0), PE-pmg (42: 9), FA (16: 3), FA (19 1), CE (18: 2), Cer (36: 1), Cer (38: 4), PC (38: 5), Cer (38: 1), and TAG (44: 3) The method of claim 1, which is selected. The at least one cancer type includes lung cancer, and the one or more lipids in the lipid set include:
The method of claim 1, wherein the method is selected from the group consisting of: The at least one cancer type includes lung cancer, and the one or more lipids in the lipid set are TAG (44: 3), PC (36: 5), PC (38: 5), Cer (38: 4). ), PE-pmg (42: 9), PC (38: 7), LysoPA (22: 0), Cer (38: 1), Cer (34: 1), and Cer (36: 1) The method of claim 1, which is selected. The at least one cancer type comprises breast cancer, and one or more lipids in the lipid set are LysoPA (22: 1), PE-pmg (42: 9), CE (20: 5), TAG (52 : 3), LysoPA (22: 0), PC (36: 3), PC (36: 4), PC (36: 2), PC (34: 2), and PC (34: 1) The method of claim 1, which is selected. The at least one cancer type includes breast cancer, and the one or more lipids in the lipid set include:
The method of claim 1, wherein the method is selected from the group consisting of: The at least one cancer type includes breast cancer and the one or more lipids in the lipid set are PC (34: 2), PC (34: 1), PC (36: 2), PC (36: 4 ), PC (36: 3), PC (38: 4), LysoPA (22: 1), PE-pmg (42: 9), LysoPA (22: 0), and CE (20: 5). The method of claim 1, which is selected. The at least one cancer type includes lung cancer and breast cancer, and the one or more lipids in the lipid set include LysoPA (22: 1), PC (36: 5), TAG (52: 3), PC (38 : 5), CE (20: 5), TAG (50: 2), BMP (39: 1), PC (34: 2), CE (18: 2), and PC (34: 1) The method of claim 1, which is selected. The at least one cancer type includes lung cancer and breast cancer, and the one or more lipids in the lipid set include:
The method of claim 1, wherein the method is selected from the group consisting of: The at least one cancer type includes lung cancer and breast cancer, and the one or more lipids in the lipid set are PC (34: 2), PC (36: 2), TAG (44: 3), CE (18 : 2), PC (34: 1), LysoPA (22: 1), PC (36: 5), Cer (36: 1), CE (20: 5), and PC (36: 3) The method of claim 1, which is selected. The method of claim 1, wherein the lipid amount is determined using mass spectrometry. The method of claim 1, wherein the amount of lipid is determined using a Fourier transform ion cyclotron resonance mass spectrometer. The method according to claim 1, wherein the sample is a processed body fluid. The sample is a processed product of body fluid, and the processed product is one or more extractions using one or more solutions containing acetonitrile, water, chloroform, methanol, butylhydroxytoluene, trichloroacetic acid, or combinations thereof. The method according to claim 1, comprising an object. The method of claim 1, wherein the prediction model includes one or more dimension reduction methods. The prediction model is selected from the group consisting of principal component analysis (PCA), shimka method (SIMCA), partial least squares discriminant analysis (PLS-DA), and orthogonal partial least squares discriminant analysis (OPLS-DA). The method of claim 1, comprising one or more methods. The method of claim 1, wherein the animal is selected from the group consisting of humans, dogs, cats, horses, cows, pigs, sheep, chickens, turkeys, mice, and rats. The at least one cancer type is carcinoma, sarcoma, blood cancer, neurological malignancy, basal cell carcinoma, thyroid cancer, neuroblastoma, ovarian cancer, melanoma, renal cell carcinoma, hepatocellular carcinoma, breast cancer, Colon cancer, lung cancer, pancreatic cancer, brain cancer, prostate cancer, chronic lymphocytic leukemia, acute lymphoblastic leukemia, rhabdomyosarcoma, glioblastoma multiforme, meningioma, bladder cancer, gastric cancer ), Glioma, oral cancer, nasopharyngeal cancer, kidney cancer, rectal cancer, lymph node cancer, bone marrow cancer, stomach cancer, uterine cancer, leukemia, basal cell cancer, cancer of epithelial cells, pyruvate 2. The method of claim 1 selected from the group consisting of cancers capable of altering the regulation or activity of carboxylase and tumors associated with any of the aforementioned cancer types. 2. The method of claim 1, comprising determining the presence or absence of more than one cancer type. A method for determining the presence or absence of at least one cancer type in an animal comprising:
-Determining the presence or absence of at least one cancer type in the animal in a predictive model by analyzing the lipid content of lipids in a lipid set in a sample derived from the animal;
Wherein the lipid content of the lipid in the lipid set comprises the input of the prediction model, and the sample comprises a body fluid or processed product thereof,
Method.